what do you want see in leadwerks 3.0

[dreamhead]

hi guys

i think of now you now lw 3 that is in the works and i am very excited to get my hands on it,but there i tinges i hope that will be in the eingine like

 

{primitive shapes that i can use for Placeholder}

{That i can choose between bakend light render(and not a out side programme) and realtime light}

{a beter editor for the schaders and a good overview of all the schaders that you can use}

{and a beter workflow for grafische designers and programmers}

and so guys what your wish for lw 3?

[Thomas]

TRIANGLES!

[Canardia]

AntiAlias is one of the most important features, since it makes programmer's art also look good.

 

Cascaded Shadow Maps is also very important, since the current maximum range of 50 meter for shadows is quite ridiculous.

 

Distance Blurred Shadows should be also in LE 3, since in reality a shadow is the more blurred the further away it is from a obstacle where the shadow is cast upon.

 

Bullet Physics would be good to have too, since its used in movies like 2012. It doesn't have much features, but it's damn fast and good for mass physics. I hope this can be done using the plug-in Driver technology of LE 3.

 

Graphicsless mode would be also good, since I want to use SDL for graphics Window creation, since I need to use SDL anyway for force feedback controllers, and of course I need a graphicsless mode for the game server program also.

 

Maybe the whole renderer could be exchanged with a Voxel renderer also, if its also based on the Driver tech.

 

PNG textures would be good to have, since DDS is quite useless unless it's uncompressed, and then it eats lots of disk space.

 

Crysis like Ocean Water would be useful too.

[dreamhead]

AntiAlias is one of the most important features, since it makes programmer's art also look good.

 

Cascaded Shadow Maps is also very important, since the current maximum range of 50 meter for shadows is quite ridiculous.

 

Distance Blurred Shadows should be also in LE 3, since in reality a shadow is the more blurred the further away it is from a obstacle where the shadow is cast upon.

 

Bullet Physics would be good to have too, since its used in movies like 2012. It doesn't have much features, but it's damn fast and good for mass physics. I hope this can be done using the plug-in Driver technology of LE 3.

 

Graphicsless mode would be also good, since I want to use SDL for graphics Window creation, since I need to use SDL anyway for force feedback controllers, and of course I need a graphicsless mode for the game server program also.

 

Maybe the whole renderer could be exchanged with a Voxel renderer also, if its also based on the Driver tech.

 

PNG textures would be good to have, since DDS is quite useless unless it's uncompressed, and then it eats lots of disk space.

 

Crysis like Ocean Water would be useful too.

 

mmm you got somes great features there Lumooja,hope the wil come true

[L B]

AntiAlias is one of the most important features, since it makes programmer's art also look good.

 

Cascaded Shadow Maps is also very important, since the current maximum range of 50 meter for shadows is quite ridiculous.

 

Distance Blurred Shadows should be also in LE 3, since in reality a shadow is the more blurred the further away it is from a obstacle where the shadow is cast upon.

 

Bullet Physics would be good to have too, since its used in movies like 2012. It doesn't have much features, but it's damn fast and good for mass physics. I hope this can be done using the plug-in Driver technology of LE 3.

 

Graphicsless mode would be also good, since I want to use SDL for graphics Window creation, since I need to use SDL anyway for force feedback controllers, and of course I need a graphicsless mode for the game server program also.

 

Maybe the whole renderer could be exchanged with a Voxel renderer also, if its also based on the Driver tech.

 

PNG textures would be good to have, since DDS is quite useless unless it's uncompressed, and then it eats lots of disk space.

 

Crysis like Ocean Water would be useful too.

What can I say - I'm stomped. For once, I would have said the exact same thing.

If this is in LE3, I'll roll on the floor crying of joy.

 

Oh wait - My 2 cents would be a river editor / multiple water planes and streamed terrain (but this is already planned).

[L B]

EDIT: Another feature I'd die for is terrain carving, like in CryEngine where you can extract parts of the terrain. Of course it's not a heightmap anymore, but it makes environments much, much more realistic.

 

Example: http://img14.imageshack.us/img14/4086/cryengine003.png

(I know these cliffs are models, but it gives the point).

[dreamhead]

EDIT: Another feature I'd die for is terrain carving, like in CryEngine where you can extract parts of the terrain. Of course it's not a heightmap anymore, but it makes environments much, much more realistic.

 

Example: http://img14.imageshack.us/img14/4086/cryengine003.png

(I know these cliffs are models, but it gives the point).

nice one

[Marcus]

I personally want to see more 3rd party libraries implemented. Josh had mentioned that it wouldn't be too dificult for him to raise the capital to hire people. I would rather that capital be spent on implementing better 3rd party libraries.

 

1 - Umbra

2 - Scaleform

3 - Purelight (included with the license. Perhaps a special deal could be made with Purelight Tech)

4 - RakNet

[Canardia]

I think it would be more useful to make some tutorial how to include 3rd party libs in your LE game.

It's really very simple if you keep it organized. You just put all the libs you want under the same directory tree, for example c:\lib\cpp\raknet, and add all their include directories to the VC++ system include so you don't ever have to think about them again, and then just drag the .lib file with the mouse into your game project.

[dreamhead]

I think it would be more useful to make some tutorial how to include 3rd party libs in your LE game.

It's really very simple if you keep it organized. You just put all the libs you want under the same directory tree, for example c:\lib\cpp\raknet, and add all their include directories to the VC++ system include so you don't ever have to think about them again, and then just drag the .lib file with the mouse into your game project.

yea for you is it easy,your a programer ,but for us the graphical artist guys .we scream and run when we see a lik of code!

[Josh]

Josh had mentioned that it wouldn't be too dificult for him to raise the capital to hire people. I would rather that capital be spent on implementing better 3rd party libraries.

I'd like to note that all capital has a cost that is proportional to risk. When your code is still in the R&D phase, the risk, and thus the cost, is much higher. Therefore although capital is available, the cost at which I could get it right now is unacceptable. You'd have to kiss streaming terrain goodbye, because it would be a new area of research I couldn't accurately estimate. You'd also run the risk of the company being shut down simply because of an error in predicting development time. Or maybe overall game sales decline and the investors get nervous. Or for no reason at all. It's common knowledge that with venture capital, 9 out of 10 of their investments fail to meet their required rate of return. A 90% chance of failure is not something I want to gamble Leadwerks on. Plus, it's easy to end up owning less than 10% of your own company after a couple rounds of funding. You'll also have a group of lawyers in Palo Alto deciding what features you should get, and it would just be my job to carry out what they tell me to do. Nothing I'm saying is controversial; That's just the way it is.

 

It seems to me like this situation creates an environment of adverse selection: A company with little to lose will be more attracted these kinds of deals. Maybe I don't know much about business, but if 90% of your investments are failing, I think there is a big disconnect between entrepreneurs and investors. It would be fun to make a new social networking site or search engine and try to get some money to build it. But gambling Leadwerks like that? I have no problem with an investor's need to make money, above all else. What I do have a problem with is the possibility of the company being shut down for reasons entirely out of my control, even if the financial potential is still good.

 

What's really scary is a lot of investors follow what's "trendy". The way they look at it is "SEO is hot right now. We should have an SEO company in our portfolio". There's an assumption that there will be one big "winner" in the field, and they want to be the one who finds it. If you have a website that does something with Google search results right now, investors will be all over you. It's great if your particular field is "hot" at the moment, but what happens when it goes out of fashion? It also indicates to me decisions are being made on a superficial basis rather than a good understanding of the technology and market.

 

Add to that the fact I believe we are near the top of the second internet bubble. Twitter has raised more than $100 million by providing a free website one smart person could have programmed. That means they are expecting to make a return of more than a billion dollars. Do I even need to explain why that will never happen?

 

Once more of the C++ core of Leadwerks Engine 3 is done, it will be possible to estimate a specific task and the returns it will generate, like an iPhone port for example. At that point the cost of capital will be lower because the risk is lower and the plan to make a return is less open-ended. You can say I need X dollars to hire Y people and make a return Z within Q months. I'd keep the costs low and pitch it to angel investors.

 

I'm going to use my own money first to get as much contracted work done as possible. I am looking for professional programmers for networking code, water, sky, and Lua game scripts.

[Wchris]

I'd like to note that all capital has a cost that is proportional to risk.

I'm going to use my own money first to get as much contracted work done as possible. I am looking for professional programmers for networking code, water, sky, and Lua game scripts.

You speak very wisely. i agree completely.

But maybe in the future something could be done to integrate as a standard the work some very skilled people did in the blogs (i think Chris Paulson IK and recast is fantastic). This could perhaps be done with lower costs and benefit everybody if documented and with official support. (just an idea or a wish, don't think you have to, i just suggest )

[Josh]

One of the reasons we are using C++ is to facilitate third party libraries. I am very much in favor of developers creating and selling add-ons, or even integrating some select features into the main engine.

 

We have a plan to allow developers to market their work and create a community economy, but I'm not allowed to talk about it yet.

[Laurens]

yea for you is it easy,your a programer ,but for us the graphical artist guys .we scream and run when we see a lik of code!

 

I agree with Lumooja on this one. Rolling all those libraries would just lead to unnecessary bloat. Josh expressed the wish to provide us with a single .lib (which I applaud) for the engine, something that will most likely not be possible either due to libraries only providing .DLL's or license restrictions prohibiting you from statically linking their libraries in a commercial product (SDL comes to mind here and I am sure there are more examples).

[Wchris]

We have a plan to allow developers to market their work and create a community economy, but I'm not allowed to talk about it yet.

i like this. it could give you some part of the opensource strenght. The capacity to federate people skills.

[Rick]

I'm going to use my own money first to get as much contracted work done as possible. I am looking for professional programmers for networking code, water, sky, and Lua game scripts.

 

1 word, RakNet. Save your money on hiring professional programmers for networking code and use something that is already tested, tried, and true. You can basically make your own RakNet wrapper to make things easier, but the underneath stuff is the important part of any networking library. Seriously, don't even waste your money on making your own. There just isn't much point in it.

[Rekindled Phoenix]

River creator just like the roads!

 

(Which has already been created, just not part of the engine natively)

[Josh]

1 word, RakNet. Save your money on hiring professional programmers for networking code and use something that is already tested, tried, and true. You can basically make your own RakNet wrapper to make things easier, but the underneath stuff is the important part of any networking library. Seriously, don't even waste your money on making your own. There just isn't much point in it.

I'll look into it. If I just compiled raknet into the engine, it would be the same as not providing any networking support, for most people. But it might be possible to use this for the internals and provide a nicer interface. I'm actually completely happy with the unreleased networking system I experimented with, except for the lack of NAT punch-through and client-side prediction.

 

AntiAlias is one of the most important features, since it makes programmer's art also look good.

Yes, this is supported with OpenGL 3 and 4.

 

Cascaded Shadow Maps is also very important, since the current maximum range of 50 meter for shadows is quite ridiculous.

CSM is already used, so I am not sure what you mean. I am interested in skewed perspective techniques, but so far have not seen any implementation that didn't have severe problems.

 

Distance Blurred Shadows should be also in LE 3, since in reality a shadow is the more blurred the further away it is from a obstacle where the shadow is cast upon.

This is a definite possibility.

 

Bullet Physics would be good to have too, since its used in movies like 2012. It doesn't have much features, but it's damn fast and good for mass physics. I hope this can be done using the plug-in Driver technology of LE 3.

I am pretty happy with Newton, but I want the addition of soft bodies and a way to handle large streaming worlds. I will take another look at Bullet, but so far I have not found anything as stable and accurate as Newton. The Penumbra games use a lot of complex joints and interacting machinery, which would not have been possible with PhysX or Havok. You take things for granted like that rolling office chair, but if you tried something like that with PhysX, it would probably explode.

 

Graphicsless mode would be also good, since I want to use SDL for graphics Window creation, since I need to use SDL anyway for force feedback controllers, and of course I need a graphicsless mode for the game server program also.

A "no graphics" renderer will be provided for making a dedicated server. You can use any of the Leadwerks renderers with an SDL window. You can replace the device input drivers with your own, based on SDL.

 

Maybe the whole renderer could be exchanged with a Voxel renderer also, if its also based on the Driver tech.

Depends on how they store they mesh data. If it's polygons, this would not be that hard. if it's some special format, it would be more difficult.

 

PNG textures would be good to have, since DDS is quite useless unless it's uncompressed, and then it eats lots of disk space.

The editor can convert PNG images to DDS in a way that is extremely convenient, and gives you full control of all conversion settings. It will even detect changes to the original png files and automatically reconvert them to DDS with the last settings used for each image. The engine will then reload the DDS textures without you telling it to. This is only in the editor, the engine itself doesn't reload stuff because that would get pretty complicated with a game running.

 

Crysis like Ocean Water would be useful too.

Since I will have a C++ source code it will finally be possible to hire outside programmers for things like this.

[Pancakes]

crysis like water needs a crysis like sky!

[Josh]

crysis like water needs a crysis like sky!

I hope we can do better than that.

[dreamhead]

and what a bout primitive shapes that i can use for Placeholder, almost every engine has it why not leadwerks?

[Laurens]

CreateCube()

CreateSphere()

 

[dreamhead]

CreateCube()

CreateSphere()

 

i mean like this

[Laurens]

Those are not primitives. That's CSG modeling. I would kill to have THAT in the editor.

[dreamhead]

Those are not primitives. That's CSG modeling. I would kill to have THAT in the editor.

yea i now and is not only great to make easy test lvl with it you can make trigers whit it too,pure awesomeness

(and didn't know it was called Constructive solid geometry )